Spatial ecology topics

Country codes: HO (Honduras), IN (Indonesia), MA (Madagascar), ME (Mexico), PE (Peru), SA (South Africa)

HO02 Dragonfly distribution in cloud forest rivers over an elevational gradient

A high diversity of dragonflies are found in rivers intersecting the cloud forested mountain top in Cusuco National Park (CNP). Including many endemics and cloud forest specialist, dragonfly communities are diverse assemblages sharing river ecosystems. From cryptic forest damselflies (genus Palaemnema) to powerful flyers (Aeshna williamsonia), this study looks into characterizing ecological niches and generating insights in the elevational distribution of dragonflies in CNP. Larval and adult stadia are collected with standardized surveys. A wide range of ecological data on the river habitat is collected and provides the base for the first ecological dissection of dragonfly communities in these mountains.

Extended Project Summary

HO03 Dung beetle ecology in the Honduran cloud forest

This topic allows students to work on one of the longest-running large-scale invertebrate ecology research projects in the Neotropics, studying the fantastically diverse dung beetles of Cusuco. The project could focus on how diversity and community structure changes over a complex matrix of elevational and habitat gradients, by adapting our existing sampling programme to set up experimental plots. There may also be the opportunity to investigate aspects of ecological genetics, or to utilise GIS in analysing local biogeography of dung beetles. Projects could involve analysing community data from the sampling programme in relation to the habitat structure measurements, or working with data from multiple teams to assess the role that dung beetles play as an indicator for forest quality or the occurrence of other species. Dung beetles also play a vital role in decomposition in the forest and in seed dispersal and the impact and effectiveness of these roles could be tested using various experimental designs. Alternatively, a project could focus on finding out more about some of the beetle species to assess how far they travel to their food source, via mark and recapture methods, or to study aspects of dung beetle ecology such as diet activity or feeding preferences.

 Extended Project Summary

HO04 Ecology of moths in the tropical cloud forest of Honduras

The moths of Cusuco are among the strangest and most beautiful in the world. This project would take advantage of the network of new high-intensity mercury vapour collecting lamps installed throughout Cusuco National Park to study the incredible diversity of moths attracted to light. Currently, two families (Sphingidae and Saturniidae) are well-studied and identifiable to species in Cusuco, but many others are also attracted to light and their diversity is poorly known. Projects could focus on establishing the diversity of the lesser known families (based on morphology or using DNA barcoding) or on increasing our understanding of the better studied species of Saturniid and Sphingid using mark recapture (for example to assess population size and dispersal, or morphological variation within and between species). There would also be scope to improve our knowledge about the process of light trapping, by studying little-known aspects such as the effects of surrounding habitat structure and the attractive radius of traps. Alternatively, studies could take advantage of collaboration with canopy access experts to undertake light trapping above the forest floor, to assess possible variation in captures and moth diversity over a vertical gradient.

Extended Project Summary

HO06 Abundance and distribution of threatened amphibian populations in Cusuco cloud forest

Amphibians are the most threatened vertebrate group globally and this is exacerbated for the amphibians of Cusuco National Park due to rapid, recent expansion of coffee farms and pastures for cattle ranching within the buffer zone and core zone of the park. If the amphibian populations continue to decline, decisions must be made regarding the value of ex-situ conservation of key species for subsequent release once the threats to the population have been resolved. In order to make such decisions, it is imperative that we have reliable estimates of amphibian population dynamics. Thus, data are urgently required on the population sizes and distributions of each of the cloud forest amphibian species and the catchments in which each occur. Data collection for this project involves sampling amphibians from both the forests and rivers at multiple locations in the park. These data may then be used to calculate reliable estimates of species abundance and may also be added to existing GIS maps of the park to investigate species distribution patterns.

Extended Project Summary

HO07 Evolution of aposematic colouration and mimicry in coral snakes

Brightly coloured and deadly coral snakes and their harmless mimics are some of the most striking denizens of Cusuco National Park. The primary driver of this type of bright coloration is convergent evolution, where natural selection impels distantly-related organisms towards a shared phenotype. Biologists have long been fascinated by how selection can cause organisms to converge on a single phenotype despite different developmental and genetic backgrounds and being separated by millions of years of evolution. Mimicry is one of the most dramatic examples of convergent evolution and in particular, coral snake mimicry is a powerful example of Batesian mimicry, which occurs when a harmless species resembles a harmful species for a protective purpose. Coral snakes are dangerously venomous elapid snakes that are usually brightly coloured and banded. Across the geographic range of coral snakes, and sometimes outside of their geographic range, harmless snakes mimic the coral snakes with the same coloured crossbands. For this project, we will study the ecological and evolutionary dynamics of coral snake mimicry in Cusuco National Forest, which is home to at least two coral snake species and nine coral snake mimicking species. Dissertation students will participate in all aspects of this project (except that venomous snakes will only be handled by a trained herpetologist), which will include 1) using spectrophotometry or full-spectrum photography to quantify color of coral snakes, mimics, and non-mimicking snakes, 2) characterizing the ecological and habitat distributions of coral snakes and mimics and 3) using plasticine models to test for predation rates on different coral snake and coral snake mimic banding patterns.

HO10 Factors affecting bird communities in the cloud forests of Cusuco

Birds are excellent indicators of forest ecosystem health as their abundance and diversity are closely related to habitat disturbance and they make ideal models because they are relatively easy to monitor and study. This topic takes advantage of the existing fixed point count survey work being undertaken for birds at over 130 survey sites across Cusuco, as well as the recently started mark-release recapture mist netting survey data. By examining species distributions and species richness across varying habitats, projects could: compare bird communities in different administrative divisions of the park (e.g. the buffer/core zones that differ in degrees of wildlife preservation and human activity); study the impact of differing disturbance levels on bird communities; investigate the impact of habitat type on bird community composition; or look at the effect of altitude on bird composition. By using covariates such as temperature, habitat structure and forest type, threshold limits for the different species could be elucidated which may have interesting implications for the impact of habitat alteration (e.g. by deforestation) in the future.

Extended Project Summary

HO12 Using camera traps to quantify human disturbance of large mammal species, Honduras

Large mammals, despite their size, are rarely observed in forest habitats and are often under-represented in biodiversity studies. By using indirect survey techniques to increase detectability, a total of 23 large mammal species have been recorded in Cusuco National Park using field signs such as footprints or droppings. These include the endangered Baird’s tapir and species which are commonly hunted for bushmeat such as red brocket deer and white collared peccaries. Camera traps are deployed throughout Cusuco National Park, placed either within 20m of the sample route network or up to 300m away from the sample routes. This enables us to examine the distribution of large mammals throughout the park with respect to distance from the park boundary, human habitation and nearby deforested patches and also distance from our transect network, focusing on the effect of human disturbance. For key target species for which there are >10 detections throughout the season, the Random Encounter Model (REM) may be employed to estimate probable abundance. Data from previous years will be available for comparison enabling temporal trends in detections to be assessed. NOTE: this project involves hiking the entire transect network and also considerable distances off transect. The park has an average slope of 30°. Thus, moderate to high levels of physical fitness are essential for students undertaking this project.

Extended Project Summary

MA35 Spatial behavioural ecology of the Malagasy giant hognose snake

The Malagasy giant hognose snake (Leioheterodon madagascariensis), is Madagascar’s largest colubrid snake, attaining sizes greater than 1.5m in length. This species has been documented engaging in ritual combat and active nest defence, and a preliminary investigation suggests that the behavioural ecology of L. madagascariensis is more complex than previously thought. For this project all sightings will be recorded using a GPS receiver and all animals encountered will be captured, measured, weighed and microchipped to allow individual identification. Other novel methods may also be employed to investigate the daily habitat usage patterns of each individual. All data collected will be visualised and analysed utilising GIS software.

Extended Project Summary

MA43 Species distribution modelling in Madagascar

Distribution models allow a set of spatial records for a given species (from our databases) to be integrated with maps of environmental covariates (e.g. elevation, climate and land cover) in order to construct and validate a statistical model of the probability that a given species will be found in a particular landscape unit. These models can then be expressed as a habitat suitability map. It will be possible for students to join one of the science teams and contribute to collecting field data for lemurs, forest birds, wetland birds, or reptiles and amphibians and then use our entire dataset to make models for a set of species using either general linear modelling (GLM) or Maxent. Outputs from these studies would be very helpful as the maps produced can feed directly into our systematic conservation planning process and inform the management of the Mahamavo region. High quality maps are also excellent communication tools for explaining the significance of the site to decision makers.

Extended Project Summary

MA44 Landscape ecology in Madagascar

By conducting biodiversity surveys we build up a knowledge base concerning patterns in the environment. However, in order to make resilient conservation plans for a dynamic future characterised by land cover change, climate change, human population growth and infrastructure development, we need to be able to understand the processes which are affecting the distribution and density of species within the landscape. It would be possible to join the teams conducting field surveys of lemurs, forest birds or reptiles to contribute to data collection, then return to base camp and use our full database, linked to our spatial data, to infer population processes from patterns of biodiversity. In particular it would be very useful to test to what extent various species in a particular guild are affected by patch size, edge effects, isolation and compactness and therefore predict the likely consequences for biodiversity of habitat fragmentation in future environmental scenarios.

Extended Project Summary

MA45 Community ecology in Madagascar

Which processes (including habitat and ecological interactions) structure communities of forest birds, reptiles and lemurs in Mahamavo? In terms of habitat, there is scope for comparison of primary and secondary dry forest and exploration of the effects of gradients in moisture between relatively moist and highly xeric forests. This might permit the identification of indicator species for particular forest types. A more sophisticated approach would be to use Mantel tests to test a suite of competing hypotheses about the environmental processes which explain pairwise dissimilarity in the community of reptiles/birds/lemurs. Pairs could be studied and differences investigated as a function of distance, difference in environmental variables such as moisture, and difference in habitat configuration. Additionally it would be possible to test whether ecological interactions, especially competition, within a taxonomic group may be structuring the community. This could be achieved by co-occurrence tests or generalised dissimilarity models. For some groups, development of ecological dissimilarity (ED) based monitoring indicators for environmental condition which track communities through ecological space through time would be a very promising direction to investigate. Alternative directions to take might be to make distribution models and then maps of betadiversity or to use numerical classification to make maps of community types. Finally, for individual taxonomic groups such as birds, it is possible to test for nestedness of communities among a set of sites.

Extended Project Summary